Access to and utilization of prenatal care services in the Unified Health System of the city of Rio de Janeiro, Brazil Acesso e utilização de serviços de pré-natal na rede SUS do município do Rio de Janeiro, Brasil
Rosa Maria Soares Madeira DominguesI Maria do Carmo LealII; Zulmira Maria de Araújo HartzIII Marcos Augusto Bastos DiasIV Marcelo Vianna VettoreV Oswaldo Cruz Foundation/ Evandro Chagas Clinical Research Institute – Rio de Janeiro (RJ), Brazil.
I
II Oswaldo Cruz Foundation/ Sérgio Arouca National School of Public Health/ Department of Epidemiology and Quantitative Methods – Rio de Janeiro (RJ), Brazil.
Nova de Lisboa University Institute of Hygiene and Tropical Medicine – Lisboa – Portugal.
III
Fernandes Figueiras Institute – Rio de Janeiro (RJ), Brazil.
IV
Brazilian Ministry of Health State Civil Servants’ Hospital – Rio de Janeiro (RJ), Brazil.
V
Corresponding author: Rosa Maria Soares Madeira Domingues. Instituto de Pesquisa Clínica Evandro Chagas/ Fiocruz. Avenida Brasil, 4365 – Manguinhos, CEP: 21040-360 – Rio de Janeiro (RJ) Brazil. E-mail:
[email protected] Financial support: PAPES/Fiocruz (process 403578/2008-6) and FAPERJ (Program Doctoral Candidate Internship Abroad, process n. 100.293/2011) Conflict of interests: nothing to declare.
Abstract Prenatal care consists of practices considered to be effective for the reduction of adverse perinatal outcomes. However, studies have demonstrated inequities in pregnant women’s access to prenatal care, with worse outcomes among those with lower socioeconomic status. The objective of this study is to evaluate access to and utilization of prenatal services in the Sistema Único de Saúde (SUS – Unified Health System) in the city of Rio de Janeiro and to verify its association with the characteristics of pregnant women and health services. A cross-sectional study was conducted in 2007-2008, using interviews and the analysis of prenatal care cards of 2.353 pregnant women attending low risk prenatal care services of the SUS. A descriptive analysis of the reasons mentioned by women for the late start of prenatal care and hierarchical logistic regression for the identification of the factors associated with prenatal care use were performed. The absence of a diagnosis of pregnancy and poor access to services were the reasons most often reported for the late start of prenatal care. Earlier access was found among white pregnant women, who had a higher level of education, were primiparous and lived with a partner. The late start of prenatal care was the factor most associated with the inadequate number of consultations, also observed in pregnant adolescents. Black women had a lower level of adequacy of tests performed as well as a lower overall adequacy of prenatal care, considering the Programa de Humanização do Pré-Natal e Nascimento (PHPN – Prenatal and Delivery Humanization Program) recommendations. Strategies for the identification of pregnant women at a higher reproductive risk, reduction in organizational barriers to services and increase in access to family planning and early diagnosis of pregnancy should be prioritized. Keywords: Pregnancy. Prenatal care. Health services accessibility. Equity in access. Program evaluation. Unified Health System.
953
Rev Bras Epidemiol 2013; 16(4): 953-65
Resumo
Introduction
A assistência pré-natal é composta por práticas consideradas efetivas para a redução de desfechos perinatais negativos. Entretanto, estudos têm demonstrado iniquidades no acesso das gestantes aos cuidados pré-natais, com piores resultados para mulheres de menor nível socioeconômico. O objetivo deste estudo é avaliar o acesso e a utilização dos serviços de pré-natal na rede SUS do Município do Rio de Janeiro e verificar sua associação a características das gestantes e dos serviços de saúde. Foi realizado um estudo transversal, no período 2007-2008, por meio de entrevista e análise de cartões de pré-natal de 2.353 gestantes em atendimento em serviços de pré-natal de baixo risco do SUS. Foi feita análise descritiva das razões referidas pelas mulheres para início tardio do pré-natal e regressão logística hierarquizada para identificação dos fatores associados à utilização do pré-natal. A ausência de diagnóstico da gravidez e dificuldades de acesso aos serviços foram as razões mais relatadas para o início tardio do pré-natal. Verificou-se acesso mais precoce de gestantes de cor branca, com maior escolaridade, primigestas e com companheiro. O início tardio foi o fator mais associado ao número inadequado de consultas, também verificado em gestantes adolescentes. Mulheres de cor preta apresentaram menor adequação na realização de exames, bem como menor adequação global do pré-natal, segundo parâmetros do Programa de Humanização do Pré-natal e Nascimento (PHPN). Estratégias para identificação de gestantes de maior risco reprodutivo, redução de barreiras organizacionais nos serviços e ampliação do acesso ao planejamento familiar e ao diagnóstico precoce da gravidez são prioritárias.
Prenatal care coverage has significantly increased in recent years, in Brazil. According to data from the DATASUS (Unified Health System Data Processing Department), the number of pregnant women without any prenatal care consultations decreased from 10.7% in 1995 to only 2% in 2009 (http://www2. datasus.gov.br/DATASUS/index.php, accessed on June 16th, 2011). In the same period, the number of pregnant women with seven or more consultations rose from 49.0% to 58.5%. It is believed that prenatal care can contribute to more favorable perinatal outcomes and there is scientific evidence of the effectiveness of certain practices routinely used in the follow-up of pregnant women1. These effects would be more pronounced in developing countries and in more socially disadvantaged populations, where factors associated with adverse outcomes such as infections, nutritional problems and exposure to harmful substances (tobacco, alcohol and other drugs) are more frequent2. However, despite the increase in prenatal care and governmental initiatives that could contribute to better perinatal results3, health problems associated with pregnancy, delivery and newborn care are still present. Among the causes of death in children aged less than one year that were considered to be preventable by SUS (Unified Health System) interventions4, those associated with prenatal care were the only ones which did not decrease between 1997 and 2006. Brazilian studies have identified problems related to prenatal care, such as late start of treatment, an inadequate number of consultations and incomplete performance of recommended procedures, which could explain the persistence of adverse outcomes5. Some of these studies specifically assessed prenatal care quality according to maternal socioeconomic characteristics and observed inequalities in this type of care, with poorer results among women with a lower income and level of education, who are precisely the ones at the highest risk of experiencing adverse outcomes and who would primarily benefit from prenatal care in theory 6-11.
Palavras-chave: Gestação. Assistência prénatal. Acesso aos serviços de saúde. Equidade no acesso. Avaliação de programas e projetos de saúde. Sistema Único de Saúde.
Rev Bras Epidemiol 2013; 16(4): 953-65
954
Access to and utilization of prenatal care services in the Unified Health System of the city of Rio de Janeiro, Brazil Domingues, R.M.S.M. et al.
Although social inequalities in health contexts are mainly determined by factors such as poverty and housing and working conditions, adequate health services can contribute to the reduction in such inequalities through their role in health promotion and disease prevention and treatment12. A representative study conducted in the city of Rio de Janeiro in the late 1990s identified a high coverage of prenatal care and the persistence of unfavorable perinatal outcomes9. In addition, it was observed that the lower the level of education and income of pregnant women, the higher the “low birth weight” and “perinatal mortality” outcomes13. There was evidence of low adequacy of prenatal care, according to the modified Kotelchuck index, and prenatal care service use (time of start of treatment and number of consultations received) was associated with several socioeconomic, demographic and biological characteristics of pregnant women9. After almost a decade, the present study aimed to assess the access to and use of prenatal care services of the SUS network in the city of Rio de Janeiro, RJ, Brazil, according to the recommendations of the Ministry of Health’s Programa de Humanização do PréNatal e Nascimento (PHPN – Prenatal and Delivery Humanization Program)3, which includes other parameters apart from the time of start of prenatal care and number of consultations, observing its association with factors related to pregnant women’s characteristics and health services.
Methods A cross-sectional study was conducted with pregnant women cared for in the SUS health network of the city of Rio de Janeiro, between November 2007 and July 2008. Two-stage cluster sampling was performed. In the first stage, health establishments providing low-risk prenatal care were selected. In the second stage, pregnant women cared for in each of the selected services were chosen. Primary selection units were stratified per type of unit as follows: Primary Health Centers (PHC), Hospitals/Maternity Hospitals, Delivery Centers
(DC) and Family Health Units (FHU). A simple random selection of eligible health units was performed in the city’s ten Administrative Areas (PA) for the PHC and Hospital strata, maintaining the same proportional distribution of units per PA in the sample that existed at the time of the study. With regard to the DC stratum, the only unit present in the city was included in this study. FMUs were selected in areas where the family health strategy has been more widely spread and those that most successfully met the criteria established by this study were included (not to be situated in an area with a high risk of violence, to have a higher number of teams available, and to have had a longer length of time of service). The “adequacy of prenatal care service” outcome was considered to define the sample size, estimated to be 50% with a significance level of 5%. A margin of error of 2.5% was defined for the PHC, hospitals and DC, while that for the FHU was 5.2%. A correction was made for the finite population and the design effect, which was estimated to be 1.5. The final sample totaled 2,187 women in the PHC, Hospital and DC stratum and 230 interviews in the FMU. As there were few refusals, the estimated sample size was obtained, totaling 2,422 pregnant women. All pregnant women cared for in the health services selected were considered to be eligible for this study, regardless of their age, place of residence and gestational age. Interviews were conducted with pregnant women and their prenatal record cards were reviewed to obtain the required data. A standardized questionnaire was used in the interviews, including questions about maternal characteristics and information about the health care received. The entire data collection was performed by health professionals and students who had been previously trained in their own health units. Prenatal record cards were manually copied or photocopied and relevant information was subsequently obtained, using a manual that standardized data collection and aimed at reducing measurement bias. All instruments were assessed and pretested in the pilot study. The questionnaires
Access to and utilization of prenatal care services in the Unified Health System of the city of Rio de Janeiro, Brazil Domingues, R.M.S.M. et al.
955
Rev Bras Epidemiol 2013; 16(4): 953-65
completed by the pregnant women were reviewed and codified by team members and data were stored using the Access software program and double data entry in all questionnaires. A prenatal care service use model was designed, based on the theoretical model that considers health service use to be influenced by external contextual factors, health system characteristics and individual factors12,14. These are categorized into predisposing factors (existing before the onset of the health problem), capacitating factors (individual or community-level) and health needs. The following predisposing factors were defined: “maternal age” (in years), “maternal level of education” (incomplete primary education, complete primary education, secondary education and higher), “ethnic group” (white, mixed or black) and “parity” (primiparous or not). Individual capacitating factors were categorized into “economic class” (economic classification from the Associação Brasileira de Empresas de Pesquisa (ABEP – Brazilian Association of Market Research Companies)15), “maternal paid work” (yes or no) and “marital status” (living with a partner or not), while the community-level capacitating factor was “type of health unit” (hospital, Primary Health Center, Family Health Unit or Delivery Center). The following criterion was used to determine health needs: “history of chronic disease” (arterial hypertension or diabetes mellitus) or “history of obstetric risks”, which is defined as the occurrence of any of the following situations in previous pregnancies: number of abortions (three or more), number of deliveries (four or more), number of Cesarean sections (two or more), occurrence of obstetric complications in previous pregnancies (hypertensive disease, gestational diabetes) or unfavorable outcomes (stillbirth, neonatal death, premature birth, low birth weight). The criterion established by the PHPN was adopted as the prenatal care service use measure3. As this study was conducted with pregnant women who were in different gestational stages, the minimum number of consultations and procedures that should have been
Rev Bras Epidemiol 2013; 16(4): 953-65
956
performed in six distinct gestational periods was defined. The start of prenatal care until the 16th gestational week was considered to be adequate; a minimum number of consultations per gestational age (one consultation in the first gestational trimester, two in the second and three in the third); results of first routine tests from the 22nd gestational week on and of second routine tests from the 34th gestational week on; and anti-tetanus vaccination after the 28th gestational week. In addition to the adequacy of each component, the PHPN’s global adequacy was also assessed and all the information used for this assessment was obtained from prenatal record cards. In the statistical analysis, each element of the sample was weighted by the inverse of its probability of selection and calibrated to recover the known distribution of prenatal consultations. During the analysis, the DCs were included in the FHU stratum to enable the design effect to be included16. First of all, the description of pregnant women’s characteristics and the problems reported by them when using the health services was made. Bivariate analysis was initially performed to assess the association between the characteristics of pregnant women and services and prenatal care service use, including the global adequacy of PHPN care and its components, with the application of the chi-square test to observe differences in proportions. Multivariate statistical analysis of the factors associated with prenatal care service use was performed using unconditional logistic regression, following the previously established hierarchical model17 (Figure 1). Variables with a significance level < 0.20 in the bivariate analysis were included in the model and those with a p-value ≤ 0.05 remained in the final model. The dependent variable was again the adequacy of prenatal care according to the PHPN’s criterion. However, considering the fact that the time of start of prenatal care may affect the consultations and that both may affect the performance of tests and vaccination, two additional analysis models were defined (models 2 and 3). In these models, the early start of prenatal care (until the 16th gestational
Access to and utilization of prenatal care services in the Unified Health System of the city of Rio de Janeiro, Brazil Domingues, R.M.S.M. et al.
week) and the number of prenatal consultations (adequate or not for the gestational age) were used as explanatory variables for tests and vaccination (Figure 1). The entire data analysis was conducted using the SPSS statistical software, version 16.0. This research project was approved by the Research Ethics Committees of the Escola Nacional de Saúde Pública Sérgio Arouca/ Fiocruz (Oswaldo Cruz Foundation/National School of Public Health) and City of Rio de Janeiro Department of Health. Data collection occurred after participants signed an informed consent form. Close attention was paid to guarantee the anonymity and confidentiality of all information obtained. Authors declared there were no conflicts of interest. Distal level
Socioeconomic and demographic characteristics Socioeconomic characteristics Maternal level of education
Intermediate level
Results Of all 2,422 pregnant women interviewed, 69 were excluded from the analysis as they did not have a prenatal record card (n = 25) or did not have a defined gestational age (n = 44). These women were not significantly different from the others included in the study and there was only a slightly lower proportion of primiparous women among those excluded (Table 1). The 2,353 pregnant women analyzed had a mean age of 24.6 years, of which one fourth were adolescents. Approximately one third of them had not achieved a primary education level and the mean level was 8.4 years of school. The majority of pregnant women belonged to socioeconomic class “C” and none of them was Proximal level
Outcomes
Maternal characteristics Reproductive history Parity Adequacy of the PHPN
History of obstetric risk
Tests
Maternal work Socioeconomic class Demographic characteristics
Maternal morbidity
Health care Type of unit
Start of prenatal care
number of consultations Vaccination
History of chronic disease
Age
Social Support
Ethnicity
Marital status
Model 1
Model 2
Model 3
Figure 1 - Hierarchical model for the analysis of prenatal care adequacy. Figura 1 - Modelo hierarquizado para análise da adequação da assistência pré-natal. Access to and utilization of prenatal care services in the Unified Health System of the city of Rio de Janeiro, Brazil Domingues, R.M.S.M. et al.
957
Rev Bras Epidemiol 2013; 16(4): 953-65
Table 1 - Socioeconomic, demographic, reproductive characteristics and prenatal care adequacy of pregnant women included and excluded from the analysis. Rio de Janeiro, RJ, Brazil, 2007 – 2008. Tabela 1 - Características socioeconômicas, demográficas, reprodutivas e adequação do pré-natal das gestantes incluídas e excluídas da análise. Município do Rio de Janeiro, Brasil, 2007 – 2008. Pregnant women included in the analysis (n = 2.353)
%
Pregnant women excluded from the analysis (n = 69)
%
≤ 19
554
23.6
14
20.1
20 – 34
1597
67.9
45
65.4
≥ 35
201
8.5
10
14.5
White
582
24.7
22
31.8
Mixed
1171
49.8
32
45.9
Black
542
23.0
15
22.3
Characteristics
p-value*
Predisposing factros Age (years)
0.069
Ethnicity
Yellow
45
1.9
0
–
Indigenous
13
0.5
0
–
Incomplete primary education
798
33.9
27
38.6
Complete primary education
754
32.0
21
30.1
Secondary education or higher
801
34.0
22
31.3
0.668
880
37.4
15
21.3
0.009
B
175
7.4
4
5.6
C
1645
69.9
45
65.3
D
425
18.0
17
24.3
E
106
4.5
3
4.8
0.508
Paid work
818
34.8
23
34.1
0.907
Living with a partner
1829
77.7
49
71.8
0.285
Hospital
841
35.8
26
37.1
Primary Health Center
1327
56.4
36
52.1
Family Health Unit/Delivery Center
185
7.8
7
10.8
0.839
History of chronic disease**
190
8.1
9
12.7
0.091
History of obstetric risk*** (n = 1,473)
544
36.9
18
33.4
0.522
0.619
Level of education
Primiparous Capaciting factors Socioeconomic class
Type of health service
Health Needs
Adequacy of prenatal care Adequacy of early start
1748
74.3
–
–
Adequacy of the number of consultations
1863
79.2
–
–
Adequacy of tests (n = 1,656)****
475
28.7
–
–
Adequacy of vaccination (n = 1,208)*****
662
54.8
–
–
Adequacy of global PHPN
905
38.5
–
–
*chi square test; **Arterial hypertension or diabetes; ***Only for women with previous pregnancies; ****pregnant women with 22 gestational weeks or more; *****pregnant women with 28 gestational weeks or more. *teste qui-quadrado; **Hipertensão arterial ou diabetes; ***Apenas para mulheres com gestação anterior; ****em gestantes com 22 semanas gestacionais ou mais; *****em gestantes com 28 semanas gestacionais ou mais. Rev Bras Epidemiol 2013; 16(4): 953-65
958
Access to and utilization of prenatal care services in the Unified Health System of the city of Rio de Janeiro, Brazil Domingues, R.M.S.M. et al.
classified in class “A”. Nearly 2% of the pregnant women reported they were Asian-descendants and 0.5% said they were indigenous, both of which were excluded from the subsequent analysis as they represented very small categories. Only one third of the women reported having paid work and more than 70% lived with a partner. PHCs were the primary place of health care, followed by hospitals and maternity hospitals. With regard to obstetric characteristics, 37.4% of participants were primiparous and 36.9% of those who had been pregnant were categorized as having a history of obstetric risks. With regard to history of chronic diseases, 8% reported having arterial hypertension and/or diabetes before the current pregnancy. Approximately 75% of pregnant women had an early start of prenatal care, which began until the 16th gestational week (Table 1). The main reason for the late start was being unaware of one’s pregnancy (40%), followed by difficulties in access to services (27%) and personal reasons (18%), such as being unsure whether to continue the pregnancy or not and having difficulty in informing family members about their pregnancy. Nearly one third of pregnant women reported having sought other health services before obtaining prenatal care where they were, thus delaying its beginning. In approximately 27% of cases, the change was motivated by the fact that they were not provided care in the first service they sought, while 33% were referred to other services, of which 17% were due to gestational risk and 16% for unknown reasons. The number of prenatal consultations was only adequate among women who were in the beginning of their pregnancy, achieving nearly 80% for all women interviewed (Table 1). More than 95% of all pregnant women reported that the following consultation was always set up at the end of the current one. However, almost one fifth of them had missed a consultation, of which 36% were due to problems in the health service and 30% for unknown reasons. Of all pregnant women who had missed any prenatal consultations, 13% found it difficult to set up a new consultation, of which 80% were due to service-related problems. A high number of pregnant women (95%) reported that blood and urine tests had been
requested and performed. Of all pregnant women who did not have these tests, nearly 70% reported difficulties in having them performed, of which 60% were due to servicerelated problems; 10%, personal reasons; and 30%, unknown reasons. However, the level of adequacy of tests was low, achieving a maximum value of only 41% in pregnant women with a gestational age of 28 to 33 weeks and close to 30% in those with 22 gestational weeks or more (Table 1). With regard to antitetanus vaccination, coverage was also low, totaling 54.8% of pregnant women with a gestational age of more than 28 weeks (Table 1). Considering all the PHPN parameters, only 38.5% of pregnant women were provided adequate prenatal care (Table 1). In the bivariate analysis (Table 2), the main factors associated with “early start of prenatal care” were higher level of education, living with a partner, being white, and not having had previous pregnancies, of which the two last ones had a borderline significance. The main factor associated with “adequate number of consultations” was early start of prenatal care. Other factors identified were pregnant women’s age (lower proportion among adolescents), higher level of education, living with a partner and being primiparous. History of chronic disease and socioeconomic class had a borderline significance and the highest level of adequacy was found in class “B” and in women with a history of chronic diseases. “Adequacy of tests” was closely associated with early start of prenatal care and an adequate number of consultations and its association with “being white” and “higher level of education” was also observed. In addition, “anti-tetanus vaccination” was associated with early start, an adequate number of consultations and the type of health service. In the global assessment of PHPN, level of education, work and history of chronic disease were the variables that maintained significant associations. In the multivariate analysis (Table 3), being white, having a higher level of education, living with a partner and being primiparous were positively associated with early start of prenatal care. Pregnant women who had a late start of prenatal care and pregnant adolescents had a
Access to and utilization of prenatal care services in the Unified Health System of the city of Rio de Janeiro, Brazil Domingues, R.M.S.M. et al.
959
Rev Bras Epidemiol 2013; 16(4): 953-65
Table 2 - Bivariate analysis* of factors associated to adequate prenatal care, considering the PHPN index and its components. Rio de Janeiro, RJ, Brazil, 2007 – 2008. Tabela 2 - Análise bivariada* dos fatores associados à adequação da assistência pré-natal segundo o critério PHPN e seus componentes. Município do Rio de Janeiro, Brasil, 2007 – 2008. Model 1 Model 2 Exposure/outcome Adeq. Adeq. p-value p-value consultations early Level of education Incomplete primary 72.5 76.9 education Complete primary 69.6 76.5 edcation Secondary 80.8 < 0.001 83.9 < 0.001 education or higher Work Yes 74.8 81.7 No 74.1 0.798 77.8 0.080 Socioeconomic class E 67.5 69.8 D 72.9 78.6 B 74.7 79.4 C 80.0 0.170 84.4 0.056 Age (years) ≤ 19 70.5 74.3 20 – 34 75.4 80.6 ≥ 35 77.6 0.086 81.5 0.009 Ethnicity White 77.8 81.7 Mixed 73.6 78.5 Black 71.7 0.054 77.3 0.208 Primiparous Yes 77.0 82.7 No 72.8 0.054 77.0 0.002 History of obstetric risk**** Yes 71.9 76.6 No 73.4 0.517 77.3 0.798 History of chronic disease Yes 78.9 84.5 No 74.0 0.169 78.7 0.058 Living with a partner Yes 77.1 81.6 No 64.7 < 0.001 70.5 < 0.001 Type of health unit Hospital 76.2 82.2 PHC 73.9 77.3 PHC/DC 69.5 0.265 78.5 0.178 Early start of prenatal Yes No NA Adeq. number of consultations Yes No NA
95.6 31.7
NA
< 0.001
Adeq. tests**
Model 3 Adeq. p-value VAT***
p-value
PHPN p-value
25.8
55.6
37.8
26.8
56.1
34.4
33.2
0.040
52.9
0.497
43.1
0.001
31.5 27.3
0.072
52.8 55.8
0.377
42.6 36.3
0.026
0.408
66.3 54.7 53.7 58.0
0.196
26.5 37.0 39.4 41.3
0.111
0.591
56.1 55.7 43.6
0.084
35.8 39.3 39.7
0.377
33.2 28.5 23.4
0.012
50.5 56.9 54.2
0.166
41.8 37.8 35.7
0.061
30.9 27.3
0.130
56.8 53.6
0.377
40.1 37.5
0.113
24.3 29.0
0.080
51.3 54.8
0.521
37.6 37.4
0.950
28.2 28.8
0.917
55.8 41.7
0.065
50.6 37.4
0.007
28.5 29.2
0.797
53.4 59.8
0.156
39.2 35.9
0.252
32.0 25.0 38.2
0.461
45.4 60.8 62.2
0.038
37.6 39.2 37.1
0.812
34.3 16.2
< 0.001
57.6 49.1
0.006
NA 5
35.1 9.4
